In an age when consistency between the macro and the micro were believed to indicate sure signs of God1s eternal truth, the anomalies which crop up as soon as a comprehensive, interdependent theory of consonance and harmony is attempted must have been a source of bewilderment and embarrassment,to Renaissance and Baroque theorists and musicians. Composite instruments such as human voices in concert have no difficulty in making pitch adjustments necessary to achieve consonance, but the keyboard with its twelve pre-set notes is inhibited. The history of temperament is an account of the struggle to endow the keyboard with something approaching the versatility of the human voice. German theorists and musicians were leaders in the recognition of the need to compromise between purity of interval and harmonic freedom. Part 1 of the dissertation traces the development of temperament from Schlick to Werckmeister, but concentrates on the seventeenth century. It attempts to show how temperaments became obsolete as musical vocabulary widened, and how, at least in Werckmeister's casey temperaments were assessed not only for their faithfulness to the accepted standards of consonance, but for their ability to give to certain keys certain affective characteristics. The music of several southern and central German composers is considered in varying degrees of detail. The most detailed study is of the music of Froberger, an important predecessor of, and influence upon, J. S. Bach. Almost exclusivelyt Froberger was a composer of keyboard music, and a musician most probably familiar with temperaments used in other musical centres of Europet because he travelled widely as a celebrated keyboard artist. As the research progressed, a need to assess in objective terms the qualities of different temperaments when applied to the keyboard music of the time became increasingly insistent. Part 2 of the dissertation shows a method of solving this problem. The psychological effects of musical sound received by the human ear are considered. By dissecting a piece of music into its smallest interacting parts, i. e. its intervals and their duration, objective comparison can be made between the piece played in one temperament and played in another or other temperaments. The complete process is laborious, but is greatly alleviated by transferring data to a suitably programmed computer, which can then process the data in relation to any temperament applied. Some of the music of Froberger has been examined in this way and conclusions drawn. The final chapter looks back over the ground covered, outlines some issues which still require investigation, and makes suggestions for further research, using the information, arguments and techniques to be found in this dissertation.